# Baseline Magnetic Resonance Imaging Assessment of Circumferential Resection Margin Predicts Long-term Survival in Rectal Adenocarcinoma: Experience from a Tertiary Care Center

## TL;DR

Baseline MRI can predict long-term survival in rectal cancer patients, even after additional treatments.

## Contribution

Baseline MRI-MRF status is shown to robustly predict long-term survival outcomes in rectal adenocarcinoma.

## Key findings

- Baseline MRI-MRF status is a strong predictor of overall and disease-free survival.
- MRI T-MRF status significantly predicts overall, disease-free, and locoregional recurrence-free survival.
- Path-CRM-positive status affects overall and disease-free survival but not locoregional recurrence-free survival.

## Abstract

In rectal adenocarcinoma, the diagnostic accuracy of baseline MRI for predicting circumferential resection margin (CRM) is established. However, data regarding the role of baseline and post-neoadjuvant chemoradiotherapy (NACTRT) MRI-mesorectal fascia (MRI-MRF)-positive status in predicting long-term oncological outcomes is relatively scarce and heterogeneous. The objective of the study is to evaluate the long-term oncological survival outcomes of baseline and post-neoadjuvant chemoradiation (NACTRT) MRI-MRF as predictors of long-term survival outcomes, i.e., overall survival (OS), disease-free survival (DFS), and locoregional recurrence-free survival (LRFS). Single center retrospective analysis from a prospectively maintained database. Patients undergoing curative surgery for rectal adenocarcinoma either upfront or post-NACTRT between July 2013 and April 2014. Patients with cT3/cT4 or N + received NACTRT before surgery. The pre-NACTRT MRI was recorded as MRI 1-MRF and post-NACTRT MRI was recorded as MRI 2-MRF. MRI scans done at presentation irrespective of further treatment were labeled as MRI T-MRF. Out of 254 patients, 217 were eligible for analysis. The median follow-up duration is 132 months. Seventy-six percent of patients received NACTRT. Overall, recurrences were seen in 68/217 (31.3%) patients, with 18 local and 50 distant recurrences. Eighty-six (39.6%) deaths were recorded, most due to disease progression. The 5-year OS of the cohort was 69.1% (95% C.I 63–75.8); 5-year DFS was 67.4% (95% C.I 61.2–74.3); and the 5-year LRFS was 91% (95% C.I 87–95.2). MRI T-MRF status was significantly associated in predicting OS, DFS, and LRFS. MRI 1-MRF status is a strong predictor for OS and DFS. The MRI 2-MRF status is a weak predictor for OS and is not associated with DFS and LRFS. The path-CRM-positive status is a significant predictor of OS and DFS, however not for LRFS. Baseline MRI-MRF status is a robust and strong predictor of long-term survival outcomes (OS, DFS, LRFS). Patients with baseline MRI-CRM-positive status have poorer outcomes irrespective of neoadjuvant therapy and poor histology features.